oncology program with 3D conformal RT; SBRT dose 20-60 Gy (5-22.5. Gy/fraction (fx)), and external beam radiation therapy (XRT) dose 45-80.5. Gy (1.5-5 ...
S604
International Journal of Radiation Oncology Biology Physics
province of Ontario. A unique feature was the development of a one page algorithm for quick and easy reference to guide practitioners to use appropriate diagnostic tests and to make early referral to the appropriate specialist for the suspicion of lung cancer. Conclusions: A useful one page algorithm has been designed for PCPs regarding patient management once lung cancer is suspected. The algorithm is the result of reviewing the best available evidence modified by local practice patterns based on experts in the management of lung cancer and relevant practitioner feedback. This algorithm is designed to be used in conjunction with available DAPs to coordinate and facilitate patient care. Author Disclosure: Y.C. Ung: None. L. Del Giudice: None. S. Young: None. E. Vella: None. M. Ash: None. P. Bansal: None. A. Robinson: None. R. Skrastins: None. R. Zeldin: None. C. Levitt: None.
3026
3025 Thoracic Reirradiation in Patients With Stereotactic Body Radiation Therapy (SBRT) as First or Second Course of Treatment J.M. Kilburn, J.G. Kuremsky, A.W. Blackstock, W.T. Kearns, C.J. Hampton, W.H. Hinson, A.A. Miller, W.J. Petty, and J.J. Urbanic; Wake Forest University School of Medicine, Winston Salem, NC Purpose/Objective(s): Limited modern data exists on management of infield failures after thoracic radiation. The need to offer patients locoregional control and prevent distant progression must be balanced against the potential risk of increased toxicity. Specifically, the role of re-irradiation with SBRT as an initial or second course of treatment is poorly defined. We review our experience treating patients with thoracic radiation therapy to previously treated high dose regions. Materials/Methods: Patients (pts) were treated as part of a thoracic oncology program with 3D conformal RT; SBRT dose 20-60 Gy (5-22.5 Gy/fraction (fx)), and external beam radiation therapy (XRT) dose 45-80.5 Gy (1.5-5 Gy/fx). 4D CT was used to define the ITV with 5 mm PTV expansions beginning in 2007. Otherwise a 10 mm PTV was applied. 8 patients prior to 2006 were treated using homogeneous tissue density for dose calculation. Pt failures were scored regarding local (at the site treated), regional (nodal disease in the hilum /mediastinum), or distant. Central versus peripheral was defined per RTOG 0813 criteria, and toxicity per NCI CTCAE v4.0. Kaplan Meier estimates were performed for local control (LC), regional (RF), distant failure (DF), and overall survival (OS). All event times were based on date of retreatment. Results: Twenty-eight pts were treated with repeat thoracic radiation therapy (2001-2011) and SBRT as a component of care. Median followup was 13 months (m) (1-52 m). Pts were included if re-irradiation was in the prior 30 Gy isodose line and intent was definitive control. Twenty pts had XRT followed by SBRT, 6 pts SBRT then SBRT, 2 pts SBRT then XRT. The median re-irradiation dose and fraction was 50 Gy and 5 fx (Range 20-70.2 Gy, 1-35 fx). The median interval between courses was 18 m. Average patient age was 63 years (45-80) with a median ECOG PS of 1. Thoracic only disease was treated in 22 pts while 6 pts had metastatic disease at retreatment; 25 pts with lung cancer (4 SCLC), and 3 pts metastatic carcinoma. Average tumor size was 2.4 cm (0.9-5.4). Median OS was 22 m, with 2 year LC at 62% (95% CI 38.1-87.3). 4 pts had RF (1 isolated). Of 22 pts with chest only disease, 2 year distant metastatic free survival was 50.9%. Ten0 pts suffered chronic grade 2-3 toxicity (6 chest wall pain, 3 dyspnea, 1 esophagitis), and 1 pt suffered grade 5 toxicity with an aorta-esophageal fistula 6 m after 54Gy in 3 fx for a central tumor after 74 Gy XRT 1 year prior. Median time to toxicity was 6 m (2-20 m). Conclusions: In field failures after thoracic radiation present a unique challenge. This analysis suggests reasonable loco-regional control and freedom from distant metastases can be established with re-irradiation using SBRT or delivering fractionated XRT in the high dose region of prior SBRT. Chronic and high grade toxicity occurred in nearly 1/3 of patients. Author Disclosure: J.M. Kilburn: None. J.G. Kuremsky: None. A.W. Blackstock: None. W.T. Kearns: None. C.J. Hampton: None. W.H. Hinson: None. A.A. Miller: None. W.J. Petty: None. J.J. Urbanic: None.
Dose Escalation Using PET Imaging for NSCLC While Maintaining RTOG 0915/0813 Criteria K.P. Orcutt,1 J.M. Larner,1 J. Buatti,2 P.W. Read,1 and K. Wijesooriya1; 1 University of Virginia, Charlottesville, VA, 2University of Iowa, Iowa City, IA Purpose/Objective(s): Despite the dramatic improvements with the advent of SBRT for the treatment of lung cancer, local tumor failure is still observed in approximately 10% of patients. Molecular imaging studies such as 18F-FDG PET reveal significant tumor heterogeneity. Although the underlying basis of the heterogeneity (perfusion, cell density, hypoxia and proliferation) is not fully understood, it is likely that heterogeneity seen in PET may predict therapeutic resistance. We therefore determined how much dose escalation could be achieved in Planning Target Volume (PTVs) sub-volumes with high standard uptake values (SUV) without violating the RTOG 0915/0813 dosimetric constraints. Materials/Methods: PET CT and CT simulation scans for 10 tumors ranging between 2cc to 20cc were used for this dosimetric study. SUVmax uptake values for each tumor were obtained from pretreatment PET CT using the Velocity system and %SUV max dose rings corresponding to values (40%SUVmax to 90%SUVmax) were defined. Pinnacle treatment planning system was used to guide the dose gradient within the PTV dose rings such that the highest uptake areas receive the highest dose using standard inverse planning constraints and non-coplanar beam angle arrangements. Dose escalation to each of the SUV uptake rings were attempted until a minor deviation in RTOG 0915/0813 criteria occurred. These SUV based dose painted plans were then compared to optimized non-dose painted plans. Results: The average mean dose for the 60-100% SUV volume can be increased up to 107.4 (12.3 STD) (range 89.4 - 124.8) Gy for a prescription dose of 60Gy while meeting the RTOG 0915/0813 dosimetric criteria. The V50% dose was increased up to 66.91 (6.40 STD) Gy where the V50% dose is defined as the dose to 50% of the PTV volume. The limiting dosimetric criteria was R50 and in some cases D105% volume as well. Conclusions: By employing 18F-FDG %SUVmax uptake rings as a guide we have dose painted the PTV volume. Our data demonstrates the feasibility of significant PTV sub-volume dose escalation for SBRT cases without violating the RTOG 0915/0813 dosimetric criteria. Author Disclosure: K.P. Orcutt: None. J.M. Larner: None. J. Buatti: None. P.W. Read: None. K. Wijesooriya: None.
3027 MR-Predictive Assay in Preoperative Lung Cancer Therapy J.C. Grecula,1 G. Jia,1 W. Yuh,1 G. Otterson,1 P. Ross,1 M. VillalonaCalero,1 K. Shilo,1 S. Lo,2 S. Ghosh,1 and N. Mayr1; 1Ohio State University, Columbus, OH, 2Case Western University, Cleveland, OH Purpose/Objective(s): To develop, refine, and test Dynamic Contrast Enhancement (DCE) Magnetic Resonance Imaging protocols and analysis methodologies for predicting response to preoperative neoadjuvant chemoradiation therapy (PNT) and resectability in non-small cell lung cancer. Materials/Methods: Patients with stage IIIA (T1-3 N2 or T3N1) nonsmall cell lung carcinoma were prospectively studied and received 45 Gy in 25 daily fractions of conformal external beam radiation therapy with 2 courses of cisplatin (50 mg/m2 IV days 1 and 8; 29 and 36) and etoposide (50 mg/m2 IV days 1-5 and 29-33). Three MRI studies (pre-contrast and 3 Tesla DCE MRI) before, during (2 weeks into treatment) and after PNT (study 1, 2, and 3 respectively) were performed for each patient. Results: Five patients have been enrolled to date. One patient progressed and developed pericardial, pulmonary artery branch, and pulmonary vein invasion and did not undergo resection. One patient was found to be unresectable at the time of surgery due to invasion of the descending aorta. Both of the unresectable patients have expired. Three patients were resectable: 2 of 3 had complete pathological response with no evidence of residual tumor at primary site or nodes; one had residual tumor at both the nodes and primary. Among the pharmacokinetic parameters (Amp, kep, and
Volume 84 Number 3S Supplement 2012 kel), kep of the second MR appeared to differentiate unresectable (nZ 2; kepZ 4.3 min-1) from resectable tumors (nZ 3; kepZ 2.3 min-1); and the case with residual tumor (n Z 1; kepZ 2.6 min-1) from those without residual tumor (n Z 2; kepZ 2.1 min-1). kep represents the exchange rate constant between blood plasma and extravascular extracellular space, which reflects the wash-in slope of the time-signal intensity curve. Conclusions: This early data shows the feasibility of performing prospective functional tumor imaging timed with the ongoing radiation course, and suggests that therapy responsiveness and resectability may be associated with lower kep. The ability to predict unresectability with the DCE-MRI’s, would not only spare patients the morbidity of unsuccessful surgery, but also eliminate the detrimental treatment gap in the patient’s therapy, which results when patients are found to be unresectable after preoperative therapy and require completion of the radiation therapy course. Supported by NCI R21 CA121582 and NCI P30 CA16058. Author Disclosure: J.C. Grecula: E. Research Grant; National Cancer Institute. G. Jia: None. W. Yuh: None. G. Otterson: None. P. Ross: None. M. Villalona-Calero: None. K. Shilo: None. S. Lo: None. S. Ghosh: None. N. Mayr: None.
3028 Assessment of Local Control Following PET/CT Guided Adaptive Radiation Therapy for Locally Advanced Non-small Cell Lung Cancer T. Altoos, M. Surucu, A. Sethi, J. Erstad, J. Magliari, and B. Emami; Loyola University Medical Center, Maywood, IL, IL Purpose/Objective(s): To evaluate whether adequate long term local control can be achieved using a novel adaptive radiation therapy for lung cancer that monitors and takes advantage of tumor regression during the course of radiation therapy using PET/CT imaging. Materials/Methods: Thirteen patients with locally advanced or inoperable non-small cell lung cancer were selected for this study. CT and PET were used to delineate the initial target volumes and organs at risk. It was not feasible to deliver doses greater than 50 Gy to the PTV due to excessive toxicity either to the lung or adjacent organs at risk using 3DCRT for all patients. Therefore, after delivering 40 to 50 Gy, patients were re-scanned to evaluate their response to radiation (and concurrent chemotherapy). The tumor volumes were re-contoured on the new CT/PET images as detailed in Emami et al. A new treatment plan using the updated target volumes was developed for each patient. As a result, the initial volume received an average dose of 46.5 Gy and the new PTV was given an additional average dose of 19.5Gy. The changes in GTV and PTV volumes were calculated. Long term local control was also evaluated by reviewing the follow-up CT scans. Results: There was significant reduction in treatment volumes when comparing the initial and re-scan volumes: 69% and 61% reduction for GTV and PTV, respectively. If the initial volumes were given an average dose of 65.9 Gy, the mean lung dose would have been 16.4 Gy. But with re-planning, the mean lung dose was reduced to 14.4 Gy. The mean doses for other organs at risk (esophagus, heart and spinal cord) were also reduced. Moreover, one year after completion of radiation treatment, radiographic local control was 62% . Nine of the 13 patients developed distant metastases at an average of 6.7 months after treatment and two patients remained disease free 5 years after completion of their radiation treatment. Conclusions: For patients with advanced lung cancer, it is often not possible to treat the pre-treatment target volume with doses considered to be curative due to intolerable doses to surrounding normal lung tissue and organs at risk. Adapting the treatment delivery, by taking advantage of reduction in tumor size during treatment, allows the delivery of prescription dose to the tumor while maintaining OAR doses below tolerance. In this study, favorable local control outcomes were achieved with this image guided adaptive radiation therapy technique. Future goals are to assess the location of the local recurrence and evaluate whether the reduction in target volume may have compromised local control in some patients.
Poster Viewing Abstracts S605 Author Disclosure: T. Altoos: None. M. Surucu: None. A. Sethi: None. J. Erstad: None. J. Magliari: None. B. Emami: None.
3029 A Clinical Guide to Evaluate Non-Monte Carlo Treatment Plans in Lung SBRT A. Sethi, I. Rusu, J. Zhung, E. Melian, and S. Nagda; Loyola University Medical Center, Maywood, IL Purpose/Objective(s): Monte Carlo (MC) treatment plans provide the most accurate estimate of dose in lung SBRT. However, most treatment planning systems (TPS) do not permit real time MC dose calculations. This leads to significant under dosing of lung targets. We evaluate planned dose distribution for lung SBRT patients and recommend a method to create realistic treatment plans using non-MC planning systems. Materials/Methods: Tissue equivalent spherical targets of different size were simulated in central and peripheral lung. Target volumes ranged from 0.47 to 66.5 cc. First, all treatment plans were calculated with a pencil beam (PB) dose algorithm to deliver 50Gy in 5 fractions to the periphery of PTV, using 10-12 non-coplanar conformal fields. To determine actual dose delivered, MC dose calculations were performed without changing beam parameters (MLC shape and monitor units). PTV dose distribution was evaluated using Dmin, Dmean, Dmax and D90. Furthermore, MC calculated spatial dose distributions within PTV were analyzed as a function of target size. To improve target coverage, the effect of increasing PTV to MLC margin on dose parameters was investigated. PTV margins were increased in steps (1-5mm) and MC calculations performed without changing beam parameters. All treatment plans were evaluated using RTOG 0813 dose criteria for D2cm, R50% and conformality index (CI). Results: Compared to PB plans, average values of PTV Dmin, Dmean, Dmax, and D90 in MC plans were lower by 219%, 108%, 64%, and 146% respectively. Spatial dose distribution within PTV indicated that the greatest dose deficit occurred in the superficial target layers. For the PTV core (5mm below surface) all dose parameters were in good agreement with delivered dose. Based on MC results, the ratio of delivered to calculated dose for Dmean, Dmax and D90 for PTV core was 974%, 954%, and 982% respectively. With an increase in PTV to MLC margin, the agreement between PB results and actual dose improved significantly. The margin increase was specific to target size, location and surrounding medium. For central lesions surrounded by lung, a uniform expansion of 5mm was found to be sufficient. Peripheral targets, adjacent to chest-wall or mediastinum, required a non-uniform expansion: 5mm within the lung, and 1-2 mm elsewhere. With this recipe, PB predictions for Dmean, Dmax and D90 were all within 2% of delivered dose. The average values of CI, D2cm and R50% were 8%, 5%, and 28% larger than PB predictions. However, except for R50%, all met RTOG specified guidelines. Conclusions: We have presented a novel and simple method of evaluating and optimizing non-MC plans so that planned dose distributions agree with delivered dose for lung SBRT patients. Author Disclosure: A. Sethi: None. I. Rusu: None. J. Zhung: None. E. Melian: None. S. Nagda: None.
3030 Incidental Mediastinal Nodal Radiation During Stereotactic Radiation for Lung Tumors: Implications for High-dose Treatment of Locally Advanced Disease S. Kwong, T. Djemil, H. Vaghefi, S. Cherian, G.M. Videtic, and K.L. Stephans; Cleveland Clinic, Cleveland, OH Purpose/Objective(s): Stereotactic radiation has been demonstrated to be safe for the treatment of both central and peripherally located stage I nonsmall cell lung cancer (NSCLC), but has to date not been applied to patients with more locally advanced disease. We sought to determine the levels of incidental hilar and mediastinal nodal radiation delivered during lung SBRT to begin to understand what might be safe dosing levels for these nodal stations.
